Keep Stave's Claude and Codex provider runtimes in sync when editing one of them. Use when a change touches `electron/providers/claude-sdk-runtime.ts`, `electron/providers/codex-sdk-runtime.ts`, `electron/providers/codex-app-server-runtime.ts`, executable lookup, env construction, CLI session launch env, tooling status probes, or any provider-agnostic behavior. Trigger on phrases like "provider runtime", "claude adapter", "codex adapter", "sibling adapter", "PATH env", "CLAUDE_CONFIG_DIR", "CODEX_HOME", "tooling status", "어댑터 대칭", "양쪽 프로바이더 확인".
72
88%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is an excellent skill description that clearly defines a narrow, well-scoped task (keeping two specific provider runtimes in sync), provides explicit trigger conditions with both file paths and natural language phrases, and is highly distinctive. The inclusion of specific file names, environment variables, and multilingual trigger terms makes it exceptionally precise for skill selection.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions and artifacts: keeping runtimes in sync, executable lookup, env construction, CLI session launch env, tooling status probes, and names specific files like 'claude-sdk-runtime.ts' and 'codex-sdk-runtime.ts'. | 3 / 3 |
Completeness | Clearly answers both 'what' (keep Claude and Codex provider runtimes in sync) and 'when' (explicit 'Use when' clause listing specific files and scenarios, plus a 'Trigger on phrases' clause with concrete keywords). | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms including file paths, conceptual phrases ('provider runtime', 'claude adapter', 'codex adapter', 'sibling adapter'), environment variables ('PATH env', 'CLAUDE_CONFIG_DIR', 'CODEX_HOME'), and even Korean-language triggers for multilingual users. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive — targets a very specific niche of keeping two named provider runtimes in sync within the Stave project, referencing exact file paths and environment variables that are unlikely to overlap with other skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a strong, well-structured skill that provides highly actionable guidance for maintaining provider runtime symmetry. Its main strengths are the comprehensive checklist with specific file paths and the clear verification steps that distinguish probe success from actual runtime success. Minor weaknesses include some verbosity in the introduction and failure modes sections, and the monolithic structure could benefit from splitting detailed sub-checklists into referenced files.
Suggestions
Trim the introductory paragraph — Claude doesn't need the motivation for why divergence is bad; start directly with the AGENTS.md rule references.
Consider extracting the 'Common Failure Modes' section into a referenced file since it partially overlaps with the checklist and adds length without new actionable steps.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is mostly efficient and well-targeted, but includes some sections that could be tightened — the introductory paragraph explaining why divergence is bad is unnecessary context for Claude, and the 'Common Failure Modes' section partially restates what the checklist already covers. The 'Use This Skill When' / 'Do Not Use When' sections are useful but slightly verbose. | 2 / 3 |
Actionability | The skill provides highly concrete, actionable guidance: specific file paths to check, a detailed checklist with precise grep targets and field names, exact verification commands (`bun run typecheck`), and a structured output format. This is an instruction-only skill that achieves full actionability without needing code examples. | 3 / 3 |
Workflow Clarity | The workflow is clearly sequenced through a numbered checklist (behavior parity → runtime options → event emission → CLI env → probe mismatch → SDK upgrade), followed by explicit verification steps that include both providers and real turn-start paths. The probe vs runtime mismatch section serves as a critical validation checkpoint, and the guardrails section provides clear feedback loops for error prevention. | 3 / 3 |
Progressive Disclosure | The content is well-organized with clear sections and a reference table, and it cross-references sibling skills (`the-ipc-contract-audit`, `the-provider-router`, `the-terminal-surface-guard`). However, with no bundle files provided, the skill is a monolithic ~150-line document; the detailed checklist items and common failure modes could potentially be split into referenced files for better progressive disclosure. The reference to `docs/providers/codex-upgrade-checklist.md` is good but is the only external content reference. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
e275ad5
Table of Contents
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.